Circular road signs recognition with soft classifiers

In this paper a computer system for recognition of the circular prohibition and obligation road signs is proposed. Its main purpose is to assist a driver with information on passing signs, which in connection with an active cruise control can prevent dangerous traffic situations. Thus, the system can help to increase safety on our roads. The proposed system consists of two main parts: a detector and a classification module. Both employ soft classifiers. The detector does colour segmentation with a support vector machine, operating in a one-class mode. Then the circular shapes are found and passed to the classifiers. The classification module is built in a form of two committee machines, each composed of a series of expert neural networks and an arbitration unit. The two machines has the same internal structure, however they operate in different input spaces. The first one works in the spatial domain, which allows very accurate assessments of the relative vertical and horizontal shifts. The second machine operates in the log-polar representation which has the ability to represent rotations as vertical shifts. Each expert of a committee machine is realized as a Hamming neural network trained with affinely deformed set of reference road signs from the data base. Selection of a single answer from a group of experts is done by an arbitration unit which operates in the winner-takes-all mode. Additionally, arbitration has been endowed with a group support mechanism which boosts answers from a group of unanimous experts. The proposed system shows very accurate and fast response on circular road signs encountered in real traffic scenes. This has been verified by experiments which results are also presented and discussed in this paper.

[1]  Boguslaw Cyganek,et al.  Rotation Invariant Recognition of Road Signs with Ensemble of 1-NN Neural Classifiers , 2006, ICANN.

[2]  José Manuel Pastor,et al.  Visual sign information extraction and identification by deformable models for intelligent vehicles , 2004, IEEE Transactions on Intelligent Transportation Systems.

[3]  Richard P. Lippmann,et al.  An introduction to computing with neural nets , 1987 .

[4]  Robert P. W. Duin,et al.  Support Vector Data Description , 2004, Machine Learning.

[5]  David G. Stork,et al.  Pattern Classification , 1973 .

[6]  Yali Amit,et al.  2D Object Detection and Recognition: Models, Algorithms, and Networks , 2002 .

[7]  Vojislav Kecman,et al.  Learning and Soft Computing: Support Vector Machines, Neural Networks, and Fuzzy Logic Models , 2001 .

[8]  Boguslaw Cyganek,et al.  Recognition of Road Signs with Mixture of Neural Networks and Arbitration Modules , 2006, ISNN.

[9]  William H. Press,et al.  Numerical recipes , 1990 .

[10]  Shigeo Abe,et al.  Support Vector Machines for Pattern Classification (Advances in Pattern Recognition) , 2005 .

[11]  Xiaohong W. Gao,et al.  Recognition of traffic signs based on their colour and shape features extracted using human vision models , 2006, J. Vis. Commun. Image Represent..

[12]  Robert P. W. Duin,et al.  Building Road-Sign Classifiers Using a Trainable Similarity Measure , 2006, IEEE Transactions on Intelligent Transportation Systems.

[13]  Bernhard Schölkopf,et al.  Learning with kernels , 2001 .

[14]  Vladimir N. Vapnik,et al.  The Nature of Statistical Learning Theory , 2000, Statistics for Engineering and Information Science.

[15]  William H. Press,et al.  Numerical Recipes in FORTRAN - The Art of Scientific Computing, 2nd Edition , 1987 .

[16]  George Wolberg,et al.  Image registration using log-polar mappings for recovery of large-scale similarity and projective transformations , 2005, IEEE Transactions on Image Processing.

[17]  Levent Burak Kara,et al.  An image-based, trainable symbol recognizer for hand-drawn sketches , 2005, Comput. Graph..

[18]  Boguslaw Cyganek,et al.  Road Signs Recognition by the Scale-Space Template Matching in the Log-Polar Domain , 2007, IbPRIA.

[19]  Yali Amit,et al.  2D Object Detection and Recognition , 2002 .

[20]  Patrik Floréen.  Computational Complexity Problems in Neural Associative Memories , 1992 .

[21]  Yizong Cheng,et al.  Mean Shift, Mode Seeking, and Clustering , 1995, IEEE Trans. Pattern Anal. Mach. Intell..

[22]  Vladimir Vapnik,et al.  The Nature of Statistical Learning , 1995 .

[23]  R. Fletcher.  Practical Methods of Optimization , 1988 .

[24]  Boguslaw Cyganek,et al.  Circular Road Signs Recognition with Affine Moment Invariants and the Probabilistic Neural Classifier , 2007, ICANNGA.

[25]  Syed Omer Gilani,et al.  Road sign detection and recognition using fuzzy artmap: A case study swedish speed-limit signs , 2006, Artificial Intelligence and Soft Computing.